MarpaX::Grammar::Parser gets Marpa to parse a grammar (using Jeffrey's 
code) and outputs a tree managed by Tree::DAG_Node.

That tree is compressed and passed to MarpaX::Grammar::GraphViz2, for 
rendering as SVG.

The compression mainly results from discarding info Marpa stores regarding 
offsets within the grammar at which each rule was found.

I was always worried about the internal representation (within Marpa) of 
that data, where changes must immediately affect my code which transforms 
Marpa's output into that Tree::DAG_Node tree.

So I finally found time to write a renderer for that tree, into a hashref, 
so I can much more easily proof-read the tree-handling code 
in MarpaX::Grammar::Parser.

The FAQ answers the obvious question: Why did I not use a module from CPAN? 
Well, I tried Data::Dumper, Data::Dumper::Concise (which is what I normally 
use), and Data::Dump::Streamer. But they just don't do what I want, so I 
wrote a module tailored to the task. It will ship 
in MarpaX::Grammar::Parser::Utils, as soon as I've again proof-read the new 
docs.

This matters because of the impact of these changes on how 
MarpaX::Grammar::GraphViz2 works, even though I do understand very few 
people will ever use this code.

Nevertheless, I do it because I dream it!

-- 
You received this message because you are subscribed to the Google Groups 
"marpa parser" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to